Effective Prompt Template for Minimax H3 V2V Character/Object Swapping
RecycledSpoons · reddit · 2026-08-09
To address common struggles with character/object swapping in Minimax H3 (Ref2va) video-to-video (V2V), the author shares a reliable prompt template.
The secret to a successful swap is being as descriptive as possible about the replacement subject. The template structures the prompt into key sections:
- subjectdefinitions: Defines the source video (providing camera movement, lighting, choreography) and the replacement subject (with an image and a 1-sentence detailed description like colors, materials, and shape).
- summary: States the goal of replacing the original subject with the new one.
- retentionanalysis: Outlines what to keep. Source video is partially preserved (background, camera path, motion trajectory), while the new subject is fully preserved (visual identity, colors, design details).
- detaileddescription: Details the specific actions and overall style post-swap.
- Audio: Preserves synchronized source audio and background music.
More from Multimodal
- Kling 4.0 fully available, shows markedly better motion coherence in fast-action scenes — alifcoder · 2026-10-16
- MJev-Spatial: Single-Pass Classification Beats Autoregressive VLMs on Spatial Reasoning — mindlab-research · 2026-10-16
- H-DDCM: Hadamard Codebooks Make Diffusion-Based Image and Video Compression Scalable — Technion · 2026-10-16
- ComfyUI v0.41.0 ships with Flux3 image model and MiniMax-H3 LightVAE support — Fcking_Chuck · 2026-10-16
- AI-generated short film 'The Dragon Rider' glides cinematically above the clouds — jijo6686 · 2026-10-16
- Realistic candid Midjourney portrait prompt: natural skin, lived-in bedroom light — tisch_eins · 2026-10-16